Regional Australian Indigenous communities are exposed to high levels of geogenic inorganic particulate matter (Shepherd et al., 2019), predominantly comprised of silica and iron oxide (Zosky et al., 2014). Exposure to these particles has been linked to a disproportionate burden of respiratory disease and infection (Blackall et al., 2018). However, the mechanism of this remains unclear. The aim of this study was to identify the impact(s) of real-world dusts and iron oxide on bacterial invasion in two common respiratory bacterial pathogens; NTHi and P. aeruginosa.
Our results suggest that geogenic dusts do not increase the invasiveness of NTHi or P. aeruginosa on alveolar epithelial cells or internalization in macrophages. However, both samples induced significantly higher rates of NTHi invasion in bronchial epithelial cells when compared to control and DEP. Interestingly, only the Newman sample increased P. aeruginosa invasion. DEP did not increase the invasion of NTHi or P. aeruginosa in any cell type. This suggests that the mechanism of DEP-induced airway infection is not through the same pathways as geogenic particulates, and that further work is required on traffic derived particles to understand the underlying mechanism. Nonetheless, in establishing that bronchial epithelial cells are more susceptible, we have provided evidence towards a potential mechanism to explain epidemiological evidence linking particulate exposure with bronchiectasis, bronchitis, and upper respiratory tract infections (Blackall et al., 2018; Chang et al., 2002; Shepherd et al., 2019).
Of particular interest was the difference in effect between the Kalgoorlie and Newman dust. Newman dust had a broader effect by increasing invasion of P. aeruginosa, and an additional strain of NTHi, compared to Kalgoorlie dusts. This observation led us to investigate the primary difference between the two samples; iron oxide (Zosky et al., 2014). In line with this, we saw significant increases in NTHi invasion in bronchial epithelial cells in response to haematite but not to silica. Importantly, we also tested another form of iron oxide, magnetite (Fe3O4), and found that the haematite effect was not replicated. This suggests that the Fe compound influences the response. Importantly, haematite is the most common form of iron in the regions where the samples were taken from and the likely the species present in the Newman sample.
While dust for Newman induced a broader spectrum response, it is important to recognise that dusts from Kalgoorlie still significantly increased NTHi invasion in bronchial epithelial cells. Given the ~ 19% iron oxide composition of the Kalgoorlie sample (compared to 30% in Newman) (Zosky et al., 2014), this suggests that the exposure does not require large percentages of iron oxide to drastically reduce the immune defence of bronchial epithelial cells. We conducted further tests to validate the level of haematite required to induce significant NTHi uptake in bronchial epithelial cells. From this, we confirmed that concentrations as low as 25 µg/mL induced significant bacterial internalization.
Lastly, we conducted genome associations to elucidate differences in gene expression that may explain the between strain variability in the response. However, there were no consistent differences in the presence or absence of the most common NTHi virulence genes. This is not unexpected as all strains of NTHi used were pathogenic. However, further analysis elucidated several clusters of genes associated with elevated haematite-induced invasion. Interestingly, these were mainly associated with metabolism. Specifically, peptide and nitrogen metabolism and heme acquisition. Many of these genes are known to modulate virulence and are associated with host crosstalk (Baddal et al., 2015). For example, enhanced nitrogen and NADH metabolism is hypothesised to aid in mitigation of redox imbalances by NTHi, but also nitrite and nitrate that arise as part of the host defence mechanism (Othman et al., 2014). Similarly, heme acquisition has been extensively linked to NTHi virulence and the presence of genes allowing greater heme acquisition is well correlated with increased pathogenesis (Morton, Seale, Bakaletz, et al., 2009). Thus, heme acquisition gene additions such as oppA, saPA, dppA & hbpA strongly imply that NTHi isolates with higher heme acquisition will be able to establish higher levels of infection (Morton, Seale, Bakaletz, et al., 2009; Morton, Seale, Vanwagoner, et al., 2009; Tanaka & Pinkett, 2019). Interestingly, our data suggests that these heme acquisition genes are most beneficial to bacterial invasion in the presence of iron exposure. NTHi are fastidious heme auxotrophs. This is the second line of evidence from our group that has suggested these bacteria may be able to benefit from environmental iron (Williams et al., 2020). Combined with our functional data, it suggests that these differences in genes are not critical to haematite-induced invasion but exacerbate invasion and could be linked to worse infection.
Our study confirmed but also challenged some of our previous work (Shepherd et al., 2019). In Shepherd et al.’s study, we demonstrated that geogenic particulates induced moderate levels of NTHi invasion in human airway epithelial cells (NuLi-1) (Shepherd et al., 2019) and concluded that haematite had no effect. This difference may be attributed to the particle concentration used and the cell type. Firstly, Shepherd et al. (2019) observed no effect of haematite at the 10µg/mL concentration (Shepherd et al., 2019). We demonstrate in our model that 10µg/mL is below the effective level of our most sensitive cells. Their results may have differed had they used 25µg/mL of haematite or higher. Nonetheless, the authors demonstrated that geogenic particulates upregulated NTHi invasion, even at low concentrations.
There are some limitations in our study that should be acknowledged. Whilst our data clearly outline the potential health impacts of iron-laden particulates, all studies were conducted in vitro. Further experiments should be carried out in vivo to address the limitations of in vitro models. Secondly, the epithelial cells used in this study were immortalised. As such, future studies should explore responses in primary cells.
This study confirms observations that geogenic particulates could increase NTHi invasion in the airway. Furthermore, we have established that the upper airway (bronchial vs alveolar) is more susceptible to particulate-induced alterations in invasion. We observed that NTHi invasion is upregulated in response to geogenic particles more than P. aeruginosa and that the response is likely driven by the iron content of the particles. Taken together this study presents evidence demonstrating that haematite, and geogenic particulates more broadly, can enhance respiratory bacterial infection by facilitating cell invasion. This may explain, at least in part, the higher incidence of chronic bronchitis and bronchiectasis in Indigenous Australians living in remote regions where ambient geogenic particulate levels are high.